On Jun 24, 10:36 pm, Martin Budden <[email protected]> wrote:
> Morris,
>
> thanks for your comments. I think it is fair to say we made a mistake
> and must do better next time. However I don't think that the mistake
> was that we "let people who routinely upgrade be unsuspecting beta
> users". We had a proper beta release back on the 27th May, see:
>
> http://groups.google.co.uk/group/TiddlyWikiDev/browse_thread/thread/1...
Martin,
Someone once said,
"Give me the fruitful error any time, full of seeds, bursting with its
own corrections. You can keep your sterile truth for yourself." :-)
First let me make one point. I believe that there must be a
separation between the development of the jQuery integration and the
upgrade facility included in TiddlyWiki.
The upgrade facility should be impeccable as it is used by people with
much investment in their data and content. They don't necessarily
have any interest or knowledge of development issues. They just want
to, perhaps naively, wish to have the latest enhancements whatever
they might be. Or in my case recently just upgrading TiddlySpot TWs to
2.5.0. and after several times be surprised that it was now 2.5.1. It
should be unthinkable that the upgrade should create problems not
already anticipated and a fix be readily understood, available, or
explainable.
The current development of integrating jQuery into TW is vastly
different from upgrades that we have been used to which were bug fixes
and updating core tweaks.
Since you are starting to replace normal TW core functions with jQuery
there could/will be mistakes and bugs that may affect many legacy
TWs. The upgrade facility must be isolated from the mistakes and bugs
that this will necessarily bring.
Of course this means that the question of whether 2.4.3/2.5.0 should
take a different upgrade path from 2.5.x. We are all aware of the
extra work this entails. And it is obvious that the 2.4.3/2.5.0 path
is eventually a cul de sac which is a disincentive for going down that
path. But it has some major advantages.
First it protects all legacy TW's, and there are many, from any new
jQuery development issues. And more importantly gives more or less a
free hand to those who want to take advantage of all jQuery has to
offer, within reason of course.
Secondly it means that the jQuery path need not be hampered, by at
once trying to break new ground, and yet try not to change anything
that has happened in the past.
Of course this conundrum is softened by the fact that any 2.4.3/2.5.0
upgrades that are not burning issues can wait for the eventual
complete conversion and release of the properly vetted TW-JQuery
2.5.x . Besides they may be more easily solved by some of the jQuery
capabilities anyhow.
In short, why make a rod for your own back with each and every release
as you go along when the final TW-jQuery will be a different animal
altogether anyhow. These are some of the reasons why I feel that the
normal upgrade path included in TW not be jeopardized, and jQuery
development restricted, by each and every major and minor release of
TW-jQuery 2.5.x.
Now to address FND's lament of the need for more testers, which is
valid and critically important.
You may remember my more or less emotional post about the development
of the Treeview plugin. My real concern was as much about
opportunities lost than the stated issue. I could see that with
praise, encouragement, and more importantly, engagement between
developers and users, that great things were in the offering.
Just as scientists provide engineers with the new discoveries that
allow them to make practical things for the masses, developers provide
the tools to provide practical applications for users. But they have
to be engaged and have one goal in common - to make something usable
and fulfill a need.
If the path of tempting users with the advantages of jQuery were more
real and tangible to their needs, then your testers would be chomping
at the bit. Take the time to provide them with what you have
implemented, give them working code examples to install and try in
their particular TWs. This is not a game or ego trip it is a real
need to cooperate to achieve a goal. It is more than unrelenting
uncompromising code - it is the practical use of it.
If I was King of TiddlyWiki I would have suggested that first, several
applications, like Treeview, be developed, and to use an idiom, users
would be 'suckered in', to the possibilities of jQuery. Once a few
tangible and usable jQuery applications were made available to users
through 2.5.0 you would have every capable user willing to cooperate
and look forward to helping test further developments. (separate from
normal upgrades of course;-)
This is not only good public relations but good sense. It would also
help the question of the perceived divide between developers and
users. Separate forums are necessary, considering some difference in
topics, but their goals are exactly the same. And some developer
needs to straddle the two with the goal in mind to inform, educate,
and converse in both languages.
Now for a word about documentation.
All good developers can look at code and naturally see its results in
their minds. It seems tedious to them to elaborate on the obvious.
This is not the case with most users. TiddlyWiki is notorious for
poor documentation. The language of , Oh its easy, just use function
() ... {foo. bar , para, para,} is disingenuous. It assumes other
people can see inside your mind. Clearly stated and actual working
examples, down to the last semi-colon should be routine if you really
want to step outside your own ego and really be helpful.
To this end you can have an army of testers, teach others of your
knowledge and at the same time advance the cause, which I shall repeat
once again, is for the benefit of the end result - find out what the
users need and make it easy for them to get.
Morris
On Jun 24, 10:36 pm, Martin Budden <[email protected]> wrote:
> Morris,
>
> thanks for your comments. I think it is fair to say we made a mistake
> and must do better next time. However I don't think that the mistake
> was that we "let people who routinely upgrade be unsuspecting beta
> users". We had a proper beta release back on the 27th May, see:
>
> http://groups.google.co.uk/group/TiddlyWikiDev/browse_thread/thread/1...
>
> Rather, I think the problem was that we didn't do sufficient upgrade
> testing on the beta release, so a bug was allowed through into the
> final release.
>
> Do you agree?
>
> Martin
>
> 2009/6/24 Morris Gray <[email protected]>:
>
>
>
> > On Jun 24, 6:39 pm, Martin Budden <[email protected]> wrote:
>
> >> we uploaded 2.5.1 on Monday and were planning to announce that day as
> >> well. Unfortunately a bug was found: upgrading did not work for any
> >> TiddlyWiki that contained non-ascii characters. The decision we took
> >> was:
>
> >> i) don't pull the 2.5.1 release
> >> ii) don't announce the upgrade, so we don't get a lot of people upgrading
> >> iii) find a fix as soon as possible, and do a 2.5.2 release containing the
> >> fix
>
> > Wouldn't it be a wiser choice to announce a release and let volunteers
> > check for bugs for a reasonable time before including it as the
> > upgrade.
>
> > As an upgrade unsuspecting users would expect that it has already been
> > debugged by a range of experts who have exhaustively tested it. It is
> > understandable that as many people as possible need to try a new
> > release to evaluate it in a variety of uses to debug it.
>
> > However is it good practice to let people who routinely upgrade be
> > unsuspecting beta users? The practice in the past gave users a choice
> > whether to be testers or not. The changes now are more than routine
> > bug fixes and tweaks, they are a dramatic deviation from what has gone
> > before.
>
> > There is a big difference between a new release and a routine upgrade.
> > And lacking any commercial or other pressures there is no reason why
> > the upgrade should not lag behind new beta releases by a safe degree.
>
> > Fortunately the first release, after the reasonably stable 2.5.0,
> > caused several people, that we know of, to denude their TiddlyWiki of
> > content should be a word to the wise.
>
> > I hope this criticism is considered constructive,
>
> > Morris
>
> > On Jun 24, 6:39 pm, Martin Budden <[email protected]> wrote:
> >> Ton,
>
> >> we uploaded 2.5.1 on Monday and were planning to announce that day as
> >> well. Unfortunately a bug was found: upgrading did not work for any
> >> TiddlyWiki that contained non-ascii characters. The decision we took
> >> was:
>
> >> i) don't pull the 2.5.1 release
> >> ii) don't announce the upgrade, so we don't get a lot of people upgrading
> >> iii) find a fix as soon as possible, and do a 2.5.2 release containing the
> >> fix
>
> >> We now have a fix, and are planning to do a 2.5.2 release today.
>
> >> Sorry about the confusion.
>
> >> As for the translation, I'll get an updated version of the English
> >> master version up today. There are a few minor changes compared to
> >> 2.4.
>
> >> Martin
>
> >> 2009/6/23 Ton van Rooijen <[email protected]>:
>
> >> > What happened to TW 2.5.0?
> >> > There hasn't been any announcement to the contrary, so TW 2.5.1 is
> >> > supposed to still be Beta.
> >> > On the official homepage of TW however, all of a sudden there is TW
> >> > 2.5.1.
> >> > But in the Timeline in the Sidebar there is no recent change, although
> >> > the NewFeatures-tiddler clearly has been updated.
> >> > Question-1: What's going on?
> >> > Furthermore I noticed the publication of Chinese TWs for 2.5.1, but I
> >> > haven't seen any change in "http://svn.tiddlywiki.org/Trunk/
> >> > association/locales/core/en/locale.en.js" since TW 2.4.
> >> > Question-2: Do we translaters need to create new versions, and where
> >> > then is our English master-copy, or are there no changes indeed as
> >> > compared to v2.4?
> >> > Btw. don't take me wrong: I'm not criticizing, but since I feel
> >> > responsible for 2 translations I am just curious.
> >> > Thanks in advance for your replies.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"TiddlyWikiDev" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/TiddlyWikiDev?hl=en
-~----------~----~----~----~------~----~------~--~---
